pith. sign in

arxiv: 2104.08315 · v9 · pith:JZH7WRVNnew · submitted 2021-04-16 · 💻 cs.CL

Surface Form Competition: Why the Highest Probability Answer Isn't Always Right

classification 💻 cs.CL
keywords probabilitysurfaceanswerchoicecompetitionformmultiplezero-shot
0
0 comments X
read the original abstract

Large language models have shown promising results in zero-shot settings (Brown et al.,2020; Radford et al., 2019). For example, they can perform multiple choice tasks simply by conditioning on a question and selecting the answer with the highest probability. However, ranking by string probability can be problematic due to surface form competition-wherein different surface forms compete for probability mass, even if they represent the same underlying concept, e.g. "computer" and "PC." Since probability mass is finite, this lowers the probability of the correct answer, due to competition from other strings that are valid answers (but not one of the multiple choice options). We introduce Domain Conditional Pointwise Mutual Information, an alternative scoring function that directly compensates for surface form competition by simply reweighing each option according to a term that is proportional to its a priori likelihood within the context of the specific zero-shot task. It achieves consistent gains in zero-shot performance over both calibrated (Zhao et al., 2021) and uncalibrated scoring functions on all GPT-2 and GPT-3 models over a variety of multiple choice datasets.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Exposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregation

    cs.CL 2026-04 conditional novelty 6.0

    TreeTracer aggregates stochastic LLM generations into syntax-aligned Sankey trees with contrastive inference to visualize and quantify hidden representational biases across demographic ontologies.

  2. A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks

    cs.LG 2026-03 unverdicted novelty 6.0

    iAmTime is a time-series foundation model that uses instruction-conditioned in-context learning from demonstrations to perform zero-shot adaptation on forecasting, imputation, classification, and related tasks.

  3. A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks

    cs.LG 2026-03 unverdicted novelty 5.0

    iAmTime is a hierarchical transformer-based time series foundation model that uses semantic tokens and instruction-conditioned prompts to infer tasks from demonstrations, achieving improved zero-shot performance on fo...